Abstract
Introduction
For this review of the past year's studies, we focused on potential initiating events, changes in the metabolome prior to the onset of type 1 diabetes, and the first promising prevention therapy, as well as other potential therapies. Further understanding of the preclinical phase of the disease process leading to type 1 diabetes, and the natural history of the disease postdiagnosis, has led to more studies evaluating interventions in at-risk subjects (stages 1 and 2), and new-onset (stage 3) studies. In addition, primary prevention studies have commenced in high genetic (HLA) risk subjects preceding the onset of islet autoimmunity.
Despite intensive study, the causality of the initiating events leading to autoimmunity remains unknown. Prospective virome analyses in the TEDDY study suggest that the persistence of enteroviruses may contribute to the development of autoimmunity in some cases. Changes in the metabolome have been demonstrated before, and after the onset of autoimmunity. In a very promising study conducted by TrialNet investigators, teplizumab (an anti-CD3 antibody) was the first drug shown to delay the onset of type 1 diabetes (stage 2 relatives) (1). Enhanced understanding of the potential mechanisms leading to efficacy is essential, as well as determining pathways that identify responders and nonresponders to therapy. Gene expression and immune signature studies are ongoing. Whether anti-CD3 will be effective earlier in the disease (stage 1) remains to be determined. The use of anti-thymocyte globulin (ATG) in this population is similarly promising.
Since it is becoming more evident that there is significant heterogeneity of the disease (and the likelihood of the presence of different endotypes), future studies aimed at prevention of type 1 diabetes will need to be individualized. In addition, since past studies have necessitated the conduct of large multicenter studies taking place over several years until we get answers, strong consideration should be given to doing smaller studies akin to those in the cancer field.
Sosenko JM, Skyler JS, Herold KC, Schatz DA, Haller MJ, Pugliese A, Cleves M, Geyer S, Rafkin LE, Matheson D, Palmer JP for the Type 1 Diabetes TrialNet Study Group
Wiedeman AE, Muir VS, Rosasco MG, DeBerg HA, Presnell S, Haas B, Dufort MJ, Speake S, Greenbaum CJ, Serti E, Nepom GT, Blahnik G, Kus AM, James EA, Linsley PS, Long SA
Holohan DR, Van Gool F, Bluestone JA
Li Q, Parikh H, Butterworth MD, Lernmark Å, Hagopian W, Rewers M, She JX, Toppari J, Ziegler AG, Akolkar B, Fiehn O, Fan S, Krischer JP, and the TEDDY Study Group
Vehik K, Lynch KF, Wong MC, Tian X, Ross MC, Gibbs RA, Ajami NJ, Petrosino JF, Rewers M, Toppari J, Ziegler AG, She JX, Lernmark A, Akolkar B, Hagopian WA, Schatz DA, Krischer JP, Hyöty H, Lloyd RE, and the TEDDY Study Group
Slowed metabolic decline after 1 year of oral insulin treatment among individuals at high risk for type 1 diabetes in the Diabetes Prevention Trial—Type 1 (DPT-1) and TrialNet oral insulin prevention trials
Sosenko JM1, Skyler JS1, Herold KC2, Schatz DA3, Haller MJ3, Pugliese A1, Cleves M4, Geyer S4, Rafkin LE1, Matheson D1, Palmer JP5 for the Type 1 Diabetes TrialNet Study Group
1University of Miami Miller School of Medicine, Miami, FL; 2Yale University School of Medicine, New Haven, CT; 3University of Florida Diabetes Institute, Gainesville, FL; 4University of South Florida, Tampa, FL; 5University of Washington School of Medicine, Seattle, WA
Background
The Diabetes Prevention Trial—Type 1 (DPT-1) initially assessed if oral insulin (7.5 mg daily) could delay the progression to type 1 diabetes in autoantibody-positive high-risk normoglycemic individuals. Although there was no overall beneficial effect of oral insulin, a secondary analysis suggested a lower occurrence of type 1 diabetes in those with high IAA (insulin autoantibody) titers. Subsequently, TrialNet carried out an appropriately powered repeat study. Despite no difference between treated and placebo groups in the primary study, at-risk subjects with reduced lower first-phase insulin responses (FPIR) showed delayed progression to clinical type 1 diabetes. The authors of this study combined the DPT-1 and TrialNet data to investigate if oral insulin slowed metabolic decline in high-risk individuals in both trials.
Methods
Participants in the DPT-1 oral insulin trial were islet cell autoantibody (ICA)- and IAA-positive relatives of individuals with type 1 diabetes, but with normoglycemia and FPIR. TrialNet participants in the oral insulin trial were positive for mIAA on two occasions: either ICA positive on two occasions or GAD and IA2 autoantibody positive on the same sample with at least one of those autoantibodies also positive on a different sample. The metabolic entry criteria were the same as those for the DPT-1 trial. Participants who had OGTT at a treatment duration of 1 year (0.75–1.25) were included in the analyses. In both the DPT-1 and TrialNet oral insulin trials, participants were randomized to receive 7.5 mg daily of recombinant human insulin crystals (Eli Lilly) or placebo orally. They were followed at 6-month intervals for the diagnostic surveillance of type 1 diabetes with 2-h OGTT. The Diabetes Prevention Trial-Type 1 Risk Score (DPTRS) for type 1 diabetes (a proportional hazards model scoring system taking into account glucose and C-peptide sums from oral glucose tolerance tests at 30, 60, 90, and 120 minutes, the log fasting C-peptide, age, and the log BMI) was calculated for participants in both studies with DPTRS threshold ≥6.75 chosen to define high risk. Metabolic endpoint comparison for high-risk participants in both studies was carried out at 1 year.
Results
After 1 year, the area under the curve (AUC) C-peptide increased significantly in the oral insulin treatment group in both trials (DPT-1: 3.28±0.17 ng/ml to 3.86±0.21 ng/ml, P=0.002, n=37; TrialNet: 3.84±1.51 ng/ml to 4.20±1.65 ng/ml, P=0.018, n=60) compared to placebo. The AUC ratio (AUC C-peptide/AUC glucose) was higher (P<0.05) in the oral insulin group in both trials, even after adjustment for age and DPTRS values. Among those with DPTRS <6.75 (n=192 in DPT-1; n=222 in TrialNet), AUC ratio values did not differ between the oral insulin and placebo groups. There was a protective effect of oral insulin in delaying progression to type 1 diabetes in both trials, but only in those with high-risk DPTRS (0.604 [0.412, 0.885; P=0.010]).
Conclusion
Oral insulin treatment for 1 year slows metabolic decline in individuals at high risk for type 1 diabetes.
Comment
This post hoc analysis demonstrated that oral insulin slows metabolic decline, especially in those with a higher risk (as defined by the DPTRS system) for progression to type 1 diabetes. A higher baseline risk with a more aggressive underlying autoimmune process thus may potentially influence response to therapy. Although there are many factors that may influence outcomes of antigen-based therapy, a higher dose may be associated with more immunomodulatory effects. The Pre-POINT study had previously demonstrated that oral insulin doses escalated to 67.5 mg daily showed a more favorable immune response in children with high-risk HLA genes. Based on the prePOINT findings, the GPPAD study group is currently investigating oral insulin as a primary preventive agent in children (from 4–7 months of age, and followed to 36 months of age) with high-risk HLA type 1 diabetes genes.
Autoreactive CD8+ T cell exhaustion distinguishes subjects with slow type 1 diabetes progression
Wiedeman AE1, Muir VS2, Rosasco MG2, DeBerg HA2, Presnell S2, Haas B2, Dufort MJ2, Speake S3, Greenbaum CJ3, Serti E4, Nepom GT1,4, Blahnik G1, Kus AM1, James EA1, Linsley PS2, Long SA1
1Translational Research Program, Benaroya Research Institute (BRI) at Virginia Mason, Seattle, WA; 2Systems Immunology, Benaroya Research Institute (BRI) at Virginia Mason, Seattle, WA; 3Diabetes Program, Benaroya Research Institute (BRI) at Virginia Mason, Seattle, WA; 4Immune Tolerance Network (ITN), Bethesda, MD
Background
Studies suggest that newly diagnosed type 1 patients have 30%–40% of functional residual β-cells, and that the rate of loss of β-cells after diagnosis is variable. Previous studies have demonstrated that anti-CD-3 antibodies (teplizumab) induce an exhausted or memory-type CD8+ T-cell profile, which may dictate the level of response to treatment with teplizumab. This is important since it has been hypothesized that islet-specific CD8+ T-cells lead to β-cell destruction.
Methods
The authors utilized high-dimensional mass spectrometry where major histocompatibility complex (MHC) tetramers (Tmr) loaded with a pooled set of HLA-A*0201-restricted peptides (derived from known islet-associated autoantigens) were used to identify islet-specific CD8+ T-cells in healthy controls (HC) and type 1 diabetes participants. Phenotyping of rare CD8+ T-cell population was then carried out with a new analytical technique called DISCOV-R (distribution analysis across clusters of a parent population overlaid with a rare subpopulation). Cryopreserved peripheral blood mononuclear cell samples from 20 HLA-A2+ HC and 46 HLA-A2+ type 1 diabetes subjects were then studied. Stratification of those with type 1 diabetes was done based on C-peptide levels (rapid progressors; n=14, <0.05 ng/mL C-peptide within 5 years of diagnosis and slow progressors; n=23, >0.1 ng/mL C-peptide at 5 or more years into disease). They were matched for age with healthy controls.
Results
There was heterogeneity of islet-specific CD8+ T-cells within individual subjects and common phenotypes across subjects. Twelve common phenotypes (clusters) were defined among CD8+ T-cells across all individuals with three clusters (representing >20% of the islet-specific Tmr+ cells) in more than 25% of the subjects. These three clusters had an activated transitional memory phenotype with high expression of HELIOS and CD27, a transitional memory phenotype with high CD27 expression but low HELIOS expression, and a memory exhausted–like phenotype (with high EOMES expression, intermediate TBET expression, and elevated expression of the multiple inhibitory receptors), respectively. There was no significant difference in the frequency of islet-specific CD8+ T-cells between rapid and slow progressors. However, islet-specific cells from the majority of rapid progressors were enriched for HELIOS +ve transitional memory phenotype, whereas the majority of islet-specific cells of slow progressors were enriched for the exhausted cluster. Islet-specific CD8+ T-cell phenotypes differed by disease progression rate, not age or disease duration.
Conclusion
Type 1 diabetes subjects with rapid decline in β-cell function are associated with presence of islet-specific CD8+ memory T-cells, while those with slow decline were associated with an exhausted CD8+ T-cell profile (with expression of multiple inhibitory receptors, limited cytokine production, and reduced proliferative capacity).
Comment
This complex yet novel mechanistic study highlights the known heterogeneity of type 1 diabetes and potential role of HELIOS +ve transitional CD8+ memory T-cells, and T-cell exhaustion in defining disease course (rapid vs slow progressors, respectively) and response to therapy. It remains to be determined if defining such phenotypes earlier in the disease course, and at what time point, might predict suitable candidates who are more likely to be susceptible to specific interventional therapeutic strategies.
Thymically-derived Foxp3+ regulatory T cells are the primary regulators of type 1 diabetes in the non-obese diabetic mouse model
Holohan DR1,2, Van Gool F1,2, Bluestone JA1,2
1Diabetes Center, University of California, San Francisco, San Francisco, CA; 2Sean N. Parker Autoimmune Research Laboratory, University of California, San Francisco, San Francisco, CA
Background
Regulatory T-cells or Tregs involved in maintaining self-tolerance express FOXP3 (forkhead box P3), a transcription factor responsible for Treg stability and its immunosuppressive function. Mutations in FOXP3 or Treg deletion can lead to autoimmunity in humans and mice, including type 1 diabetes. The relative roles of thymically derived Tregs (tTregs) and peripherally derived Tregs (pTregs) in controlling autoimmunity still remain unclear. pTregs, found in the large intestine lamina propria (LILP), contain the transcription factor retinoic acid–related orphan receptor-γt (RORγt). These RORγt+ Tregs are induced by the presence of complex microbiota and provide anti-inflammatory functions. RORγt+ pTregs responsive to the short-chain fatty acid (SCFA) butyrate are generated in the gut where they promote the growth of border-dwelling bacteria that protects against microbial exposure and maintain normal metabolite profile. Deletion of the conserved noncoding sequence (CNS)1 in the Foxp3 locus in mice has shown selective impairment of pTreg generation without disrupting tTreg generation. The authors of this study have attempted to elucidate the role of pTregs in the pathogenesis of type 1 diabetes in NOD mice.
Methods
The CRISPR-Cas9 genome editing technique was used to delete the CNS1 region of FoxP3 in the NOD mice and create NOD CNS1-/- mice. Based on a 705 base pair (bp) region (+ 2003 to 2707 from the transcriptional start site spanning the regulatory element of the FoxP3 gene), single guide RNAs (sgRNAs) were generated to create a CRISPER/Cas9-targeted deletion of a 736 bp sequence (+ 1976 to + 2711) spanning the CNS1 region. In vitro FoxP3 induction was carried out using TGF-B; this was followed by in vivo studies on Treg population after the prediabetic NOD mice were sacrificed.
Results
Deletion of CNS1 impaired in vitro induction of Foxp3 in naïve NOD CD4+ T-cells, but there were no alterations in Treg population in most lymphoid and non-lymphoid tissues of NOD and NOD CNS1-/- mice. Treg expression of Nrp1 and Helios (cell surface markers associated with tTregs) within the thymus were comparable between NOD and NOD CNS1-/- females. However, there was a small but significant decrease in the percentage of RORγt+ Treg populations in NOD CNS1-/- mice. CNS1 deletion did not alter the development of T1D or glucose tolerance despite increased insulitis in NOD CNS1-/- mice.
Conclusion
pTregs do not have a dominant role in controlling T1D in the NOD mouse model.
Comment
This mechanistic study highlighted the major role of tTregs in the pathogenesis of type 1 diabetes. Although the study did not find a major role of pTreg defect in disease pathogenesis, it demonstrated that CNS1 deletion led to a decrease in gut pTregs, which are important for maintaining a stable gut microbiome. This observation may suggest a potential role of CNS-dependent gut pTregs in disease pathogenesis during the initial phase that additionally might be influenced by environmental factors. This finding may provide further insight into the interactions between the immune system and environment that determines type 1 diabetes susceptibility and outcomes.
Longitudinal metabolome-wide signals prior to the appearance of a first islet autoantibody in children participating in the TEDDY study
Li Q1, Parikh H1, Butterworth MD1, Lernmark Å2, Hagopian W3, Rewers M4, She JX5, Toppari J6,7, Ziegler AG8–10, Akolkar B11, Fiehn O12, Fan S12, Krischer JP1, and the TEDDY Study Group
1Health Informatics Institute, Morsani College of Medicine, University of South Florida, Tampa, FL; 2Department of Clinical Sciences, Lund University/CRC, Skåne University Hospital SUS, Malmo, Sweden; 3Pacific Northwest Diabetes Research Institute, Seattle, WA; 4Barbara Davis Center for Childhood Diabetes, University of Colorado Denver, Aurora, CO; 5Center for Biotechnology and Genomic Medicine, Medical College of Georgia, Augusta University, Augusta, GA; 6Department of Pediatrics, Turku University Hospital, Turku, Finland; 7Research Centre for Integrative Physiology and Pharmacology, Institute of Biomedicine, University of Turku, Turku, Finland; 8Institute of Diabetes Research, Helmholtz Zentrum München, Munich, Germany; 9Forschergruppe Diabetes, Technical University of Munich, Klinikum Rechts der Isar, Munich, Germany; 10Forschergruppe Diabetes e.V. at Helmholtz Zentrum München, Munich, Germany; 11National Institute of Diabetes and Digestive and Kidney Diseases, National Institutes of Health, Bethesda, MD; 12Genome Center, University of California, Davis, Davis, CA
Background
Previous studies have suggested an association of the metabolome and lipidome of infants with development of islet autoantibodies and subsequently type 1 diabetes. These metabolites include reduced keto-leucine, elevated glutamic acid, elevated odd-chain triglycerides (TGs) and polyunsaturated fatty acid–containing phospholipids, and lower levels of methionine. However, these studies were done either at or after seroconversion to islet autoantibodies. The TEDDY study group has sought to define longitudinal metabolic patterns of TEDDY subjects taking into account effects of age and possible association with the risk of different first-appearing autoantibodies in a nested case control study.
Methods
A total of 8676 children based on HLA-DR-DQ haplogenotypes for the risk of type 1 diabetes were enrolled in the prospective study group. For this nested case control study, samples from 414 case and 1234 matched control subjects were analyzed. Blood samples were collected from 3 months of age every 3 months up to development of IA before 4 years of age. Mass spectrometry (MS) with gas chromatography time-of-flight (GC-TOF MS) and liquid chromatography quadrupole time of flight (LC-QTOF MS) were utilized to analyze for longitudinal metabolome and lipidome profiles in these nested case-control TEDDY plasma samples.
Results
Higher levels of dehydroascorbic acid (DHAA) (oxidized vitamin C) after birth (3 months of age) were significantly associated with the subsequent appearance of autoantibodies (either GADA-first, P=0.028, or IAA-first, P=0.0048). Interestingly, γ-aminobutyric acid (GABA) after birth was found to be associated with future risk of IAA-first only (P=0.0113). Isoleucine and valine were significantly lower prior to seroconversion in IAA first subjects, while proline and alpha-ketoglutarate were negatively associated with risk of GADA-first. Higher levels of unsaturated TGs, unsaturated phosphatidylcholine (PCs), and phosphatidylethanolamine (Pes) were associated with risk for both autoantibodies, while sugar alcohols and butyrates were enriched only for IAA-first, and plasmalogens, sphingomyelins, and unsaturated diglycerides were enriched only for GADA-first. For either first-appearing autoantibody, unsaturated TG and Pes were decreased prior to seroconversion.
Conclusion
Different metabolome pattern precedes the appearance of IAA-first and GADA-first in those at risk for type 1 diabetes. These include DHAA, GABA, amino acids, and fatty acids.
Comment
Changes in the metabolome are detected at different time points prior to seroconversion to IAA-first or GADA-first. These findings further contribute to the notion that the pathogenesis of the disease varies in those presenting with either IAA (early) or GADA (later) first. The majority of these metabolites are associated with either insulin synthesis or secretion. As this study only looked into either IAA-first or GADA-first, the longitudinal metabolic profile of those who develop multiple autoantibodies simultaneously at seroconversion is still unknown. The interplay of genetics and environmental factors potentially leading to these changes should be sought. Further studies are needed in this arena. Whether such insights will enable us to institute specific therapies aimed at the primary prevention (or delay in development of islet autoantibodies) of type 1 diabetes remains to be determined.
Prospective virome analyses in young children at increased genetic risk for type 1 diabetes
Vehik K1, Lynch KF1, Wong MC2, Tian X2, Ross MC2, Gibbs RA3, Ajami NJ2, Petrosino JF2, Rewers M4, Toppari J5,6, Ziegler AG7–9, She JX10, Lernmark A11, Akolkar B12, Hagopian WA13, Schatz DA14, Krischer JP1, Hyöty H15,16, Lloyd RE2, and the TEDDY Study Group
1Health Informatics Institute, Morsani College of Medicine, University of South Florida, Tampa, FL; 2Alkek Center for Metagenomics and Microbiome Research, Department of Molecular Virology and Microbiology, Baylor College of Medicine, Houston, TX; 3Human Genome Sequencing Center, Baylor College of Medicine, Houston, TX; 4Barbara Davis Center for Diabetes, University of Colorado, Aurora, CO; 5Department of Pediatrics, Turku University Hospital, Turku, Finland; 6Research Centre for Integrative Physiology and Pharmacology, Institute of Biomedicine, University of Turku, Turku, Finland; 7Institute of Diabetes Research, Helmholtz Zentrum München, Munich, Germany; 8Forschergruppe Diabetes, Klinikum Rechts der Isar, Technische Universität München, Munich, Germany; 9Forschergruppe Diabetes e.V, Munich, Germany; 10Center for Biotechnology and Genomic Medicine, Medical College of Georgia, Augusta University, Augusta, GA; 11Department of Clinical Sciences, Lund University/CRC, Skane University Hospital, Malmö, Sweden; 12National Institute of Diabetes and Digestive and Kidney Diseases, Bethesda, MD; 13Pacific Northwest Research Institute, Seattle, WA; 14Department of Pediatrics, University of Florida Diabetes Institute, Gainesville, FL; 15Department of Virology, Faculty of Medicine and Health Technology, Tampere University, Tampere, Finland; 16Fimlab Laboratories, Pirkanmaa Hospital District, Tampere, Finland
Background
Previous studies have suggested an association between enterovirus infections and islet autoimmunity. Some enteroviruses infect β-cells in vitro and have been detected in pancreatic islets of type 1 diabetes. Interestingly, β-cells strongly express cell surface Coxsackie and adenovirus receptor (CXADR) genes, which can facilitate enterovirus infection. Studies of human pancreata and cultured islets have shown significant variation in enteroviral virulence to β cells between serotypes and within the same serotype. Herein, the TEDDY study group has evaluated fecally shed viruses prior to the development of islet autoimmunity and type 1 diabetes.
Methods
The participants in this nested case-control study (matched for sex, clinical site, and family history) were selected from the TEDDY cohort of children with high-risk HLA genotypes and persistent confirmed islet autoimmunity. Stool samples (8654 stools for the islet autoimmunity and 3380 stools for the T1D nested-matched case-control studies) were collected monthly from ages 3 to 48 months and then quarterly until the age of 10 years. Viral RNA was extracted from stool samples, sequenced, underwent reverse transcription, and was amplified. These amplicons then underwent a complex virome analysis utilizing a clustering algorithm (VirMAP) that reconstructs putative viral genomes, using a mapping assembly strategy based on nucleotide and translated nucleotide alignment information.
Results
Enterovirus (EV-A or EV-B) was detected in 12.8% of the 8654 stools and in at least one stool for 55.4% of the 766 islet autoimmunity nested-matched case-control children. Sequence data from virus capsid regions enabled identification of the exact serotype of EV-B in 81.2% of positive samples. The number of stools positive for EV-B positive (each additional positive sample, OR=1.20, 95% CI=1.01–1.42, P=0.04), or consecutively positive for EV-B (OR=3.05, 95% CI=1.64–5.69, P=0.0005), were significantly associated with developing islet autoimmunity. Children with prolonged shedding of same EV-B (OR=2.50, 95% CI=1.19–5.26) or children with consecutive positive with different serotype (OR=2.18, 95% CI=0.74–6.46) had similarly higher odds of being an islet autoimmunity case compared to children negative for an EV-B. Those with prolonged shedding of Coxsackie B virus (CVB) were more likely to develop islet autoimmunity (OR=2.49, 95% CI=1.12–5.54, P=0.03); those positive for CVB4 serotype with no evidence of prolonged shedding were also more likely to develop islet autoimmunity (OR=2.75, 95% CI=1.18–6.40, P=0.02). Human mastadenovirus C (HAdV-C) was detected in fewer islet autoimmunity cases as compared to controls (OR=0.55, 95% CI=0.38–0.81, P=0.003, false discovery rate=0.03). Children who carried the minor SNP allele rs6517774 in the CXADR gene region were more likely to develop islet autoimmunity.
Conclusions
Prolonged Enterovirus B (EV-B) rather than independent, short-duration EV-B infections may be involved in the development of islet autoimmunity in some young children. Human mastadenovirus C infections early in life and CXADR rs6517774 independently correlated with islet autoimmunity negatively and positively, respectively.
Comment
The study demonstrated that prolonged shedding or consecutive shedding of EV-B was associated with islet autoimmunity. The tropism of CVB to the pancreatic islets appears to be due to the CXADR receptor (used by CV and adenovirus to infect cells), which is strongly expressed by β-cells. Children with rs6517774 and rs2824404 alleles (of CXADR) were strongly associated with islet autoimmunity, even in the absence of EV-B (low exposure to EV-B). The authors of the study have speculated if the prolonged shedding in patients who go on to develop islet autoimmunity is a reflection of defective innate immune response and consequent immune dysregulation and islet autoimmunity. This is the first time it has been shown that a variant in this virus receptor is tied to an increased risk for β-cell autoimmunity. The presence in early life of adenovirus C, a virus that can cause respiratory infections, was associated with a lower risk of developing autoimmunity. It remains to be investigated whether having adenovirus C in early life would protect from developing β-cell autoimmunity. Adenoviruses use the same β-cell surface receptor as Coxsackievirus B, which may offer one clue to explain this connection, although further research is needed to fully understand the details. It is thus possible that viral vaccines may lead to the primary prevention of type 1 diabetes in some patients at risk for islet autoimmunity.
Footnotes
Author Disclosure Statement
BN, NB, and DS have no financial conflicts/interests.